The long-range goals of this laboratory are to identify and characterize novel tumor suppressor genes, in order to facilitate their utilization for early detection, diagnosis, prognosis, and therapy of breast cancer. The specific research is based on the prior adaptation in this laboratory of subtractive hybridization designed for positive direct selection of candidate tumor suppressor genes active against breast cancer. In prior research in this laboratory, several candidate genes, expressed in normal mammary epithelial cells but not in patient-derived tumor cells, have been identified and cloned as cDNAs. They include fibronectin; several keratins including K5; NB-l (a calmodulin-related gene); caNl9, a gene that encodes a 10KD calcium binding protein in the S100 protein family; glutathione-S-transferase pi, encoding a well-known detoxification enzyme; and spr-l, a gene whose expression is induced by UV and certain chemical carcinogens; and the connexin genes Cx26 and Cx43 that encode gap junction proteins, and thereby regulate gap junction intercellular communication (GJIC). For initial investigation the connexin genes were chosen based on evidence of other laboratories that GJIC is a key mechanism in growth control and normal cellular differentiation. The hypothesis to be tested is that GJIC plays an important role in regulating growth of normal cells and that this regulatory mechanism is lost during breast tumor progression. In Specific Aim I it is proposed to examine the regulation and growth control functions of Cx26 and Cx43 in normal and in patient-derived mammary tumor cells. Expression levels of connexin mRNAs and proteins will be quantitated, as well as GJIC by a new method designed in this laboratory measuring dye transfer by cytofluorography. Cx26 and Cx43 will be transfected into tumor cells, individually and together, and the transfectants will be used for comparative studies of growth regulation individually and under conditions of GJIC formation with normal cells, and by comparative growth studies of normal/transfectant mixed cell populations in nude mice. Cell cycle regulated changes in Cx gene expression and communication will be examined using a new synchrony method applicable both to normal and to tumor cells. The upstream regulatory regions of the gene will be sequenced, and studies will be undertaken to identify binding sites and transcription factors that regulate Cx transcription. In Specific Aim II it is proposed to investigate the expression of connexin transcripts and proteins in vivo, using frozen (or where possible fixed) sections of breast tumors obtained from surgery for which coded medical and pathology records are available. The purposes are first to determine what changes in connexin expression arise during tumor growth and progression, and second to initiate the use of candidate tumor suppressor genes as markers for early detection and prognosis. When it is efficient time-wise to do so, additional candidate tumor suppressor genes will be included in certain of these studies.